CN102111255A - Adaptive error detection method used for cellular mobile communication system - Google Patents

Abstract

The invention discloses an adaptive error detection method used in a cellular mobile communication system, which comprises the following steps: performing the calculation of the time interval between two adjacent protocol data units received in a radio link control protocol layer entity _TI sampling, TI smoothing for filtering, _RTT sampling for obtaining the round-trip delay of each protocol data unit, RTT smoothing for filtering, and for obtaining each K samples of the retransmission times of error protocol data units, used to realize k smoothing of filtering; set the timeout value of the rearrangement timer. The present invention optimizes the t-Reordering value of SR-ARQ by adaptively estimating the retransmission time of the MAC PDU, thereby accurately and timely detecting the lost packet of the MAC layer, avoiding unnecessary retransmission of RLC, and reducing the retransmission time of the packet. delay. The invention has low implementation complexity and does not increase the overhead of the wireless link air interface.

Description

Adaptive Error Detection Method for Cellular Mobile Communication System

technical field

The invention belongs to the field of mobile communication, in particular to an error detection method used in a cellular mobile communication system.

Background technique

The evolution idea of the new-generation mobile communication system is to continuously enhance the data transmission capability of the air interface by adopting high-order modulation methods and various effective channel coding technologies, and finally make the wireless interface and core network all move towards IP (Internet Protocol).

Since the wireless interface part of the mobile communication network adopts wireless transmission mode, compared with the wired network, wireless communication has the characteristics of high bit error rate, which will increase the loss rate of data packets and cause sudden packet loss. A high packet loss rate will degrade the performance of high-level protocols, thereby affecting the quality of data services, especially the performance of Transmission Control Protocol (TCP, Transmission Control Protocol) at the transport layer. For this reason, the air interface of the cellular mobile communication system usually adopts an error correction code with strong error correction capability to reduce the packet loss rate of the wireless interface. However, error-correcting coding brings additional overhead, and the stronger the error-correcting capability, the more expensive this overhead is. At the same time, since the characteristics of wireless channels change with time, coding and modulation are required to change dynamically with the state of the channel. Therefore, the physical layer of the air interface of the new generation mobile communication system adopts adaptive coding and modulation technology (AMC, Adaptive Modulation) and Coding).

AMC can adapt to the dynamic change of wireless channel quality, while reducing the bit error rate of air wireless interface, it can improve the utilization rate of wireless spectrum. However, even with AMC technology, the air wireless interface of the cellular mobile communication system still has a high packet loss rate, which cannot meet the high-level protocol. For this reason, the new generation of cellular mobile communication system adopts Automatic Repeat reQuest (ARQ, Automatic Repeat reQuest) technology in the link layer of the wireless interface to further reduce the packet loss rate of the wireless interface.

In the evolution version (R5-R9) of Wideband Code Division Multiple Access (WCDMA, Wideband Code Division Multiple Access), the medium access control (MAC, Medium Access Control) layer and the radio link control protocol (RLC, Radio Link Control) layer adopts ARQ technology. Among them, the MAC layer combines the traditional forward error correction (FEC, Forward Error Correction) and ARQ technology, and the sender decides whether to retransmit through the positive confirmation/negative confirmation (ACK/NACK, ACKnowledgment/Non-ACKnowledgement) fed back by the receiver. And during retransmission, the incremental redundancy (IR: Incremental Redundancy) method is adopted to adapt to various channel conditions and improve system performance. The RLC layer adopts the selective retransmission mechanism (SR-ARQ: Selective Repeat ARQ). The receiver tells the sender the sequence number of the correctly received packet and the sequence number of the wrong packet through ACK/NACK, and the sender only retransmits the wrong packet.

The main purpose of adopting SR-ARQ at the RLC layer in the evolved version of WCDMA is to correct residual errors of Hybrid Automatic Repeat Request (HARQ, Hybrid ARQ), so as to further reduce the packet loss rate. In order to correct the residual error of HARQ, the SR-ARQ at the receiver must first determine which packets have not been correctly retransmitted in the HARQ of the MAC layer. For this reason, the SR-ARQ of RLC jointly uses out-of-sequence detection and reordering timer (t-Reordering) overtime mechanism to detect packets lost at the MAC layer, as shown in Figure 1: at time _t1 in Figure 1, the receiver (It can be user equipment (UE: user equipment) or evolved base station (eNB: evolved Node-B), and the corresponding sender is eNB or UE, the same below) A packet with a sequence number (SN: Sequence Number) of 5 is received, because The packet with sequence number 4 has been lost, so at this time, VR(R)<VR(H) (VR(R): the sequence number of the next sequential packet expected to be received; VR(H): the maximum sequence number of the received packet plus 1). When VR(R)<VR(H) is found, the receiver starts the reordering timer (t-Reordering) and waits for the HARQ retransmission of the MAC layer. At t ₂ , the retransmission timer expires, t ₂ -t ₁ That is, the value of the retransmission timer. When t-Reordering times out, the receiver sends a status report (STATUS) packet to the sender, and sets the ACK sequence number (ACK-SN) and NACK sequence number (NACK-SN) to 6 and 4 respectively. After the receiver receives the STATUS packet, Select the packet whose sequence number is 4 for retransmission, so as to realize error recovery.

The above SR-ARQ process shows that the value of the t-Reordering timer is actually the time for the RLC layer to wait for the MAC layer to retransmit lost packets through the HARQ retransmission mechanism. The value of the t-Reordering timer has a great impact on the performance of the system. The specific analysis is as follows:

1) The value of the t-Reordering timer is too small. If the value of the t-Reordering timer is set too small, it will cause the RLC receiver to send the STATUS packet in advance, that is, send the STATUS during the HARQ retransmission of the lost packet p, and the RLC sender will retransmit p after receiving the STATUS . Therefore, too small a timer value will lead to unnecessary retransmission, which will not only cause repeated packets at the receiver, but also waste channel resources and increase the energy consumption of the sender.

2) The value of the t-Reordering timer is too large. If the value of the t-Reordering timer is set too large, it will cause the RLC receiver to delay sending STATUS, so that the RLC sender cannot retransmit the lost packet in time, which will increase the delay of the packet on the air interface and the retransmission of the packet. Reassemble (reassemble) delay, and increasing the delay will affect the performance of the upper layer transmission protocol and the implementation of the business.

Contents of the invention

Purpose of the invention: for above-mentioned existing problems and deficiencies, the purpose of the present invention is to provide a kind of adaptive error detection method for cellular mobile communication system, this method can detect wireless link packet loss accurately and in time.

Technical solution: In order to realize the above-mentioned invention object, the technical solution adopted in the present invention is: a kind of self-adaptive error detection method for cellular mobile communication system, comprises the following steps:

(1) In the radio link control protocol layer entity, perform _TI sampling for calculating the time interval between two adjacent protocol data units received, and _TI smoothing for filtering, in the media access control layer RTT sampling for obtaining the round-trip delay of each protocol data unit, RTT smoothing for filtering and k sampling for obtaining the number of retransmissions of each error protocol data unit, and k smoothing for filtering ;

(2) set the timeout value of the rearrangement timer;

_TI sampling in described step (1), _TI smoothing comprise the steps:

1) T _I sampling: each time the RLC protocol layer entity receives a new RLC protocol data unit with the serial number x, it records the receiving time t _pr1 of the protocol data unit, and receives the next wireless link When controlling the sublayer protocol data unit, if the sequence number of the protocol data unit is x+1, record the receiving time t _pr2 of the protocol data unit, and calculate a sample of _TI as t _pr1 -t _pr2 ;

2) T _I smoothing: Calculate the smooth T _I value according to the following formula:

$\stackrel{\&OverBar;}{{\stackrel{\&OverBar;}{T}}_I}(\mathrm{no}+1)=\mathrm{\&beta;}\&Center\; Dot;\stackrel{\&OverBar;}{{\stackrel{\&OverBar;}{T}}_I}\left(\mathrm{no}\right)+(1-\mathrm{\&beta;})\&Center\; Dot;T_I(\mathrm{no}+1),\stackrel{\&OverBar;}{{\stackrel{\&OverBar;}{T}}_I}\left(0\right)=0,$ 0≤β≤1,

分别为n+1时刻和n时刻的平滑后的协议数据单元接收时间间隔，β为T_I的平滑因子，T_I(n+1)为n+1时刻测得的协议数据单元接收时间间隔的样本；In the formula,

are the smoothed PDU receiving time intervals at time n+1 and n, respectively, β is the smoothing factor of T _I , and T _I (n+1) is the PDU receiving time interval measured at time n+1 sample;

RTT sampling in described step (1), RTT smoothing comprise the steps:

A) RTT sampling: each time the hybrid automatic repeat request process sends a positive confirmation/negative confirmation, record the sending time t _ps of the positive confirmation/negative confirmation, and record the receiving time of the protocol data unit when receiving the next protocol data unit t _pr , calculate a sample of RTT as t _pr -t _ps , and upload the RTT sample to the hybrid automatic repeat request entity;

B) RTT smoothing: every time the HARQ entity receives an RTT sample in step A), it uses the following formula to calculate a smooth RTT value, and uploads the calculation result to the RLC protocol layer entity:

$\stackrel{\&OverBar;}{\stackrel{\&OverBar;}{\mathrm{RTT}}}(\mathrm{no}+1)=\mathrm{\&alpha;}\&Center\; Dot;\stackrel{\&OverBar;}{\stackrel{\&OverBar;}{\mathrm{RTT}}}\left(\mathrm{no}\right)+(1-\mathrm{\&alpha;})\&CenterDot;\mathrm{RTT}(\mathrm{no}+1),\stackrel{\&OverBar;}{\stackrel{\&OverBar;}{\mathrm{RTT}}}\left(0\right)=0,$ 0≤α≤1,

分别为n+1时刻和n时刻的平滑往返时延，α为RTT的平滑因子，RTT(n+1)为n+1时刻测得的往返时延的样本；In the formula,

are the smoothed round-trip delays at time n+1 and time n, respectively, α is the smoothing factor of RTT, and RTT(n+1) is the sample of the round-trip delay measured at time n+1;

K sampling, k smoothing in described step (1) comprise the steps:

a) k sampling: When the HARQ process detects a protocol data unit error for the first time, set the retransmission times k corresponding to the protocol data unit to 0, and then each time a retransmission of the protocol data unit is received , k plus 1, until the protocol data unit is successfully retransmitted or reaches the maximum number of retransmissions R _max , upload the value of k to the hybrid automatic repeat request entity;

b) k smoothing: every time the HARQ entity receives a k value in step a), it uses the following formula to calculate the smoothed k value, and uploads the calculation result to the radio link control protocol layer entity:

$\stackrel{\&OverBar;}{\stackrel{\&OverBar;}{k}}(\mathrm{no}+1)=\mathrm{\&gamma;}\&CenterDot;\stackrel{\&OverBar;}{\stackrel{\&OverBar;}{k}}\left(\mathrm{no}\right)+(1-\mathrm{\&gamma;})\&CenterDot;k(\mathrm{no}+1),\stackrel{\&OverBar;}{\stackrel{\&OverBar;}{k}}\left(0\right)=1,$ 0≤γ≤1,

In the formula, are the smooth HARQ PDU retransmission times at time n+1 and n time respectively, and k(n+1) is the number of HARQ PDU retransmission times measured at time n+1 Sample, γ is the smoothing factor of k;

Described step (2) comprises:

When the RLC protocol layer entity detects out-of-sequence PDUs, it calculates and sets the timeout value of the rearrangement timer according to the following formula:

${\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; R</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&Center\; Dot;</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; RTT</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; I</span>}}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; \&le;</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&le;</span>{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; ,</span>$

为重排定时器的超时值。In the formula,

It is the timeout value of the rearrangement timer.

Beneficial effects: the present invention optimizes the value of t-Reordering of SR-ARQ by self-adaptively estimating the retransmission time of the protocol data unit (PDU, Protocol Data Unit) of the MAC layer, thereby accurately and timely detecting the packet lost by the MAC layer, Therefore, unnecessary retransmission of RLC is avoided, and the retransmission delay of packets is reduced. The invention has low implementation complexity and does not increase the overhead of the wireless link air interface.

Description of drawings

FIG. 1 is a schematic diagram of SR-ARQ of RLC;

FIG. 2 is a schematic diagram of the interaction process between the RLC entity and the MAC entity of the sender and the receiver;

Fig. 3 is a schematic diagram of the calculation process of the t-Recording value;

Fig. 4 is a schematic diagram of a specific embodiment of the present invention.

Detailed ways

Below in conjunction with accompanying drawing and specific embodiment, further illustrate the present invention, should be understood that these embodiments are only for illustrating the present invention and are not intended to limit the scope of the present invention, after having read the present invention, those skilled in the art will understand various aspects of the present invention Modifications in equivalent forms all fall within the scope defined by the appended claims of this application.

为接收方正确接收到4号PDU的时间，因此，为避免t-Reordering过早或延后超时，在图2示例中，TR的最优值

为：As shown in Figure 2, PDU No. 4 successfully reaches the receiver after being retransmitted twice (transmitted three times) by one of the processes. Figure 2

It is the time when the receiver correctly receives the No. 4 PDU. Therefore, in order to avoid t-Reordering prematurely or delay timeout, in the example in Figure 2, the optimal value of TR

for:

${\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="T\; R"\; filenames="HDA0000043278950000011.png"\; state="\{\{state\}\}">T</figure-callout></span><figure-callout\; id="2"\; label="T\; R"\; filenames="HDA0000043278950000011.png"\; state="\{\{state\}\}">\; </figure-callout>}}_{\mathrm{<span\; class="notranslate">\; </span>}}^{\mathrm{<span\; class="notranslate">R</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="t"\; filenames="HDA0000043278950000011.png"\; state="\{\{state\}\}">t</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>$

Further, as can be seen from Figure 2,

${\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="T\; R"\; filenames="HDA0000043278950000011.png"\; state="\{\{state\}\}">T</figure-callout></span><figure-callout\; id="2"\; label="T\; R"\; filenames="HDA0000043278950000011.png"\; state="\{\{state\}\}">\; </figure-callout>}}_{\mathrm{<span\; class="notranslate">\; </span>}}^{\mathrm{<span\; class="notranslate">R</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="t"\; filenames="HDA0000043278950000011.png"\; state="\{\{state\}\}">t</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="1"\; label="t"\; filenames="HDA0000043278950000021.png"\; state="\{\{state\}\}">t</figure-callout></span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 3</span>}\mathrm{<span\; class="notranslate">\; RTT</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; I</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$

For a more general situation, a HARQ PDU is successfully retransmitted k times after an error occurs:

${\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; R</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&CenterDot;</span>\mathrm{<span\; class="notranslate">\; RTT</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; I</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; )</span>$

In formula (3), RTT is the round-trip delay of sending a PDU in the HARQ process, and T _I is the time interval between two adjacent PDUs received by the RLC entity. RTT and _TI are related to the wireless channel rate, available resource block interval, resource block size, and PDU processing time, and k is also related to the state of the wireless channel, the corresponding modulation and demodulation and coding methods. Although the samples of RTT, _TI and k can be obtained by measurement, due to the time-varying characteristics of the wireless channel and the dynamic resource scheduling algorithm, RTT, _TI and k are time-varying. Therefore, we need to smooth RTT, _TI and k by Equation (4), Equation (5) and Equation (6).

$\stackrel{\&OverBar;}{\mathrm{RTT}}(\mathrm{no}+1)=\mathrm{\&alpha;}\&Center\; Dot;\stackrel{\&OverBar;}{\mathrm{RTT}}\left(\mathrm{no}\right)+(1-\mathrm{\&alpha;})\&CenterDot;\mathrm{RTT}(\mathrm{no}+1),\stackrel{\&OverBar;}{\mathrm{RTT}}\left(0\right)=0,$ 0≤α≤1, (4)

$\stackrel{\&OverBar;}{T_I}(\mathrm{no}+1)=\mathrm{\&beta;\&alpha;}\&Center\; Dot;\stackrel{\&OverBar;}{T_I}\left(\mathrm{no}\right)+(1-\mathrm{\&beta;})\&CenterDot;T_I(\mathrm{no}+1),\stackrel{\&OverBar;}{T_I}\left(0\right)=0,$ 0≤β≤1, (5)

$\stackrel{\&OverBar;}{\stackrel{\&OverBar;}{k}}(\mathrm{no}+1)=\mathrm{\&gamma;}\&Center\; Dot;\stackrel{\&OverBar;}{\stackrel{\&OverBar;}{k}}\left(\mathrm{no}\right)+(1-\mathrm{\&gamma;})\&Center\; Dot;k(\mathrm{no}+1),\stackrel{\&OverBar;}{\stackrel{\&OverBar;}{k}}\left(0\right)=1,$ 0≤γ≤1, (6)

分别为n+1时刻和n时刻的平滑后的往返时延(平滑指通过一个低通滤波，对抽样值进行“平均”)，RTT(n+1)为n+1时刻测得的往返时延的样本，α为RTT的平滑因子。In formula (4),

are the smoothed round-trip delays at time n+1 and time n respectively (smoothing refers to "averaging" the sampled values through a low-pass filter), and RTT(n+1) is the round-trip time measured at time n+1 Delayed samples, α is the smoothing factor of RTT.

分别为n+1时刻和n时刻的平滑后的PDU接收时间间隔，T_I(n+1)为n+1时刻测得的PDU接收时间间隔的样本，β为T_I的平滑因子。In formula (5),

are the smoothed PDU receiving time intervals at time n+1 and time n respectively, T _I (n+1) is the sample of the PDU receiving time interval measured at time n+1, and β is the smoothing factor of T _I.

In formula (6), are the smoothed HARQ PDU retransmission times at time n+1 and time n, respectively, k(n+1) is the sample of HARQ PDU retransmission times measured at time n+1, and γ is the smoothing factor of k.

The above-mentioned values of α, β, and γ can be set by the user according to specific conditions.

After smoothing RTT, T _I and k, we get n time estimated value of for:

${\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; R</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&CenterDot;</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; RTT</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; I</span>}}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; \&le;</span>\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&le;</span>{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 7</span>}<span\; class="notranslate">\; )</span>$

In the formula, R _max is the maximum number of retransmissions specified by the MAC.

As shown in Figure 3, the setting process of the t-Reordering timer value is completed by the receiver, involving the receiver's MAC HARQ entity, HARQ process and RLC entity (HARQ process and HARQ entity are defined in 3GPP TS 36.321 [2] The functional entity of the MAC layer, the RLC entity is the functional entity of the RLC layer defined in 3GPPTS 36.322 [3]), including the RTT and k sampling of the HARQ process, the RTT and k smoothing of the HARQ entity, and the sampling of the _TI of the RLC entity , smoothing, and setting of the t-Reordering timer value.

The RTT sampling function of the HARQ process is used to obtain the round-trip delay of each PDU. The specific implementation process is as follows: each time HARQ sends an ACK/NACK, record the sending time t _ps of the ACK/NACK, and record the time when the next PDU is received. The receiving time t _pr of the PDU calculates a sample of RTT as t _pr -t _ps , and uploads the RTT sample to the HARQ entity.

The RTT smoothing function of the HARQ entity is to implement a simple filtering function, that is, each time an RTT sample from the HARQ process is received, the smoothed RTT value is calculated using formula (4), and the calculation result is uploaded to the RLC entity.

The k-sampling function of the HARQ process is used to obtain the number of retransmissions of each error PDU. When the HARQ receiving process detects a PDU error for the first time, the number of retransmissions corresponding to the PDU is set to 0, and then each time the PDU is received For one retransmission of PDU, add 1 to k until the PDU is successfully retransmitted or the maximum number of retransmissions is reached, then upload the value of k to the HARQ entity.

The k-smoothing function of the HARQ entity is to filter the retransmission times k, that is, each time a k sample from the HARQ process is received, the smoothed k value is calculated using formula (5), and the calculation result is uploaded to the RLC entity.

The TI sampling smoothing function of the RLC entity is used by the RLC receiving entity to calculate the time interval between two adjacent PDUs received. The specific process is as follows: each time a new RLC PDU with a sequence number of x is received, record the receiving time of the PDU t _pr1 , when receiving the next RLC PDU, if the sequence number of the PDU is x+1, record the receiving time t _pr2 of the PDU, and calculate a sample of T _I as t _pr1 -t _pr2 , and finally according to formula (6) Computes the smoothed _TI value.

)，利用式(7)计算并设置定时器t-Reordering的超时值。The main function of the t-Reordering setting function block is when the RLC detects out-of-sequence PDUs, according to the current T _I , RTT and k values (ie

), using formula (7) to calculate and set the timeout value of timer t-Reordering.

FIG. 4 is an embodiment of the present invention used in a 3GPP LTE system. The ellipse in Figure 4 represents the service access point, which is the interface between the upper and lower protocol layers in the protocol. The service data is in the logical channel DTCH (Dedicated Traffic Channel, dedicated traffic channel), DL-SCH (DownLink Shared Channel, downlink shared channel) , UL-SCH (Uperlink Shared Channel, uplink shared channel) channel transmission; logical channel DCCH (Dedicated Control Channel, dedicated control channel) is used to transmit control information; logical channel priority distinction function and multiplexing and demultiplexing functions in 3GPPTS36 .321[2] defines that it mainly completes the multiplexing and demultiplexing of logical channels (including dedicated control channels and dedicated service channels), and schedules different logical channels according to channel priorities.

In Figure 4, the four functions of this patent (including RTT, k sampling function, RTT, k smoothing function, TI sampling smoothing function, and t-Recording setting function) are respectively embedded in 3GPPTS36.321[2] In the HARQ process and HARQ entity of the MAC layer defined in 3GPPTS36.322 [3], the functional entity of the RLC layer is defined. All functions are implemented on the receiving side, which can be eNB or UE. The specific implementation is as follows:

1) RTT sampling: count the round-trip delay of the MAC PDU, which is implemented in the HARQ process of the MAC layer;

2) k sampling: count the number of retransmissions of the MAC PDU with errors, and implement it in the HARQ process of the MAC layer;

3) RTT smoothing: according to the RTT sampling value, the average value of the round-trip delay of the MAC PDU is calculated by using formula (4), and implemented in the HARQ entity of the MAC layer;

4) k smoothing: according to k sampling value, utilize formula (5) to calculate the mean value of the retransmission times of the MAC PDU of error, implement in the HARQ entity of MAC layer;

5) TI sampling smoothing: Statistical samples of the time interval between two adjacent PDUs received by the RLC entity, and calculate the average value according to the sample using formula (6), which is realized by the RLC entity in the confirmation mode (AM: Acknowledge Mode) of the RLC layer ;

6) t-Reordering setting: according to the average value of RTT, k, TI, use formula (7) to calculate the value of t-Reordering, and set t-Reordering according to the calculation result, and realize it in the AM RLC entity of RLC layer.

This patent proposes a dynamic optimization method for the PDU error detection timer (t-Reordering) in the RLC entity in the WCDMA evolution version. To avoid unnecessary retransmission of PDUs at the RLC layer, thereby avoiding waste of wireless resources, saving energy consumption, and reducing the transmission delay of RLC PDUs. The method is independently completed by the receiver, does not need to exchange signaling messages through the air interface, and is simple to implement.

Claims (1)

1. an adaptive error detection method that is used for cell mobile communication systems is characterized in that comprising the steps:

(1) in the radio link control layer entity, is used to calculate the T in the time interval of adjacent two protocol Data Units that receive respectively _ISample, be used to realize the T of filtering _ISmoothly, in media access control layer, be used to obtain the round-trip delay of each protocol Data Unit the RTT sampling, be used to realize the RTT of filtering is level and smooth and be used to obtain the number of retransmissions of each mistake protocol Data Unit the k sampling, be used to realize that the k of filtering is level and smooth;

(2) timeout value of resetting timer is set;

T in the described step (1) _ISampling, T _ISmoothly comprise the steps:

1) T _ISampling: the radio link control layer entity is whenever received the radio link control layer protocol data cell that a new sequence number is x, writes down the time of reception t of this protocol Data Unit _Pr1, when receiving next control sub layer of wireless link protocol Data Unit,, then write down the time of reception t of this protocol Data Unit if the sequence number of this protocol Data Unit is x+1 _Pr2, calculate T simultaneously _IA sample be t _Pr1-t _Pr2

2) T _ISmoothly: calculate level and smooth T according to following formula _IValue:

$\stackrel{\&OverBar;}{{\stackrel{\&OverBar;}{T}}_I}(n+1)=\mathrm{\&beta;}\&CenterDot;\stackrel{\&OverBar;}{{\stackrel{\&OverBar;}{T}}_I}\left(n\right)+(1-\mathrm{\&beta;})\&CenterDot;T_I(n+1),\stackrel{\&OverBar;}{{\stackrel{\&OverBar;}{T}}_I}\left(0\right)=0,$ 0≤β≤1，

In the formula,

Be respectively n+1 constantly and the protocol Data Unit time of reception interval behind constantly level and smooth of n, β is T _ISmoothing factor, T _I(n+1) the protocol Data Unit time of reception that records constantly for n+1 sample at interval;

RTT sampling in the described step (1), RTT smoothly comprise the steps:

A) RTT sampling: positive acknowledgement/Negative Acknowledgement of the every transmission of hybrid automatic repeat request process, the transmitting time t of record positive acknowledgement/Negative Acknowledgement _Ps, when receiving next protocol Data Unit, write down the time of reception t of this protocol Data Unit _Pr, a sample that calculates RTT is t _Pr-t _Ps, and this RTT sample is uploaded to mixes automatic repeat requests entity;

B) RTT is level and smooth: mix automatic repeat requests entity and whenever receive a steps A) in the RTT sample, just utilize following formula to calculate level and smooth RTT value, and result of calculation be uploaded to the radio link control layer entity:

$\stackrel{\&OverBar;}{\stackrel{\&OverBar;}{\mathrm{RTT}}}(n+1)=\mathrm{\&alpha;}\&CenterDot;\stackrel{\&OverBar;}{\stackrel{\&OverBar;}{\mathrm{RTT}}}\left(n\right)+(1-\mathrm{\&alpha;})\&CenterDot;\mathrm{RTT}(n+1),\stackrel{\&OverBar;}{\stackrel{\&OverBar;}{\mathrm{RTT}}}\left(0\right)=0,$ 0≤α≤1，

In the formula,

Be respectively the n+1 moment and n level and smooth round-trip delay constantly, α is the smoothing factor of RTT, the sample of the round-trip delay that RTT (n+1) records constantly for n+1;

K sampling in the described step (1), k smoothly comprise the steps:

A) k sampling: when hybrid automatic repeat request process is checked through a protocol data unit error for the first time, the number of retransmissions k of this protocol Data Unit correspondence is put 0, whenever receive the once re-transmission of this protocol Data Unit afterwards, k adds 1, retransmits success or reaches maximum retransmission R up to this protocol Data Unit _MaxAfter, the k value is uploaded to the automatic repeat requests entity of mixing;

B) k is level and smooth: mix automatic repeat requests entity and whenever receive a k value in the step a), just utilize following formula to calculate level and smooth k value, and result of calculation is uploaded to the radio link control layer entity:

$\stackrel{\&OverBar;}{\stackrel{\&OverBar;}{k}}(n+1)=\mathrm{\&gamma;}\&CenterDot;\stackrel{\&OverBar;}{\stackrel{\&OverBar;}{k}}\left(n\right)+(1-\mathrm{\&gamma;})\&CenterDot;k(n+1),\stackrel{\&OverBar;}{\stackrel{\&OverBar;}{k}}\left(0\right)=1,$ 0≤γ≤1，

In the formula,

Be respectively the n+1 moment and the n automatic repeat requests protocol Data Unit of level and smooth mixing number of retransmissions constantly, the sample of the automatic repeat requests protocol Data Unit of the mixing number of retransmissions that k (n+1) records constantly for n+1, γ is the smoothing factor of k;

Described step (2) comprising:

When the radio link control layer entity detects out-of-sequence protocol Data Unit, calculate and be provided with the timeout value of resetting timer according to following formula:

$T_R^{*}\left(n\right)=(\stackrel{\&OverBar;}{k}\left(n\right)+1)\&CenterDot;\stackrel{\&OverBar;}{\mathrm{RTT}}\left(n\right)-2\stackrel{\&OverBar;}{T_I},1\&le;\stackrel{\&OverBar;}{k}\left(n\right)\&le;R_{\max },$

In the formula, Attach most importance to and arrange the timeout value of timer.

Images