CN101828903B - Linear detection method for signals in non-free space environment around human body - Google Patents

Abstract

A linear detection method of a signal in a non-free space environment around a human body in the field of signal processing technology, comprising the following steps: preprocessing a received wireless signal, and starting timing; respectively using an existing linear detection method to process the received signal For preliminary detection, the linear detection method with the priority of the detection performance curve is used as the initial detection method; when the human ECG signal meets the physical realization conditions, the initial detection method is used as the optimal detection method to detect the received signal; otherwise, the energy detection method is used as the first detection method. The optimal detection method detects the received signal; during the detection process of the received signal using the optimal detection method, when the posture of the human body changes or the time of the timer is greater than the time threshold, the detection is performed again. The invention effectively avoids the problem that a single detection method cannot cope with various channel conditions, improves the screening efficiency of the detection method, and improves the detection performance in the non-free space around the human body.

Description

A linear detection method for signals in non-free space around the human body

technical field

The invention relates to a method in the technical field of signal processing, in particular to a linear detection method for signals in a non-free space environment around a human body.

Background technique

Wireless devices are developing towards miniaturization and personalization, among which the development of wireless ad hoc networks is very representative. In many wireless application fields, medical wireless technology has been developed very rapidly. In order to achieve the purpose of real-time medical monitoring, it is necessary to place sensors in many parts of the body and process the data measured by these sensors in a timely manner. So the concept of Wireless Body Area Network (WBAN) is proposed. In the application of wireless body area network in the medical field, it is necessary to monitor the physiological information of the human body through wireless devices placed on the surface of the human body, and the body area network performs summary processing. Since it is working near the human body, parameters such as signal transmission power in the wireless body area network are constrained. This puts forward higher requirements on the signal detection capability of the entire wireless body area network system. For signal detection in a general wireless free space environment, many different linear detection methods (including energy detection, cyclic spectrum detection, matched filter detection, etc.) have been proposed.

Considering the highly random nature of human motion, human motion estimation is added as an aid throughout the detection process. In terms of motion estimation, Tanmay Pawar et al. elaborated on the use of human ECG in their "Transition Detection in Body Movement Activities for Wearable ECG (Transition Detection in Body Movement Activities in Portable ECG Detection System)" published in IEEE Magazine. (Electrocardiographic signal) signal is used as a method of human motion estimation, demonstrating the feasibility of using human physiological signals to detect human motion; The channel characteristics of the human body)" specifically studied the influence of human body movement and posture changes on the transmission signal, and explained the important influence of human body posture on the channel around the human body.

In fact, the wireless signal detection environment around the human body is very different from the standard free space. The standard free space refers to a region that can be described by a large-scale or small-scale mathematical model, has the same isotropic properties along each axis, and has a homogeneous structure. Wireless signals in free space satisfy linear additivity, which can be described by appropriate mathematical models. Since each sensor node in the wireless body area network is placed on the surface of the human body, the wireless channel between them is seriously affected by the movement and posture of the human body. Although the wireless channel in the body area network is also affected by the surrounding environment and the characteristics of the antenna itself, the posture of the human body plays a very important role. Because for a wireless channel composed of any two sensor nodes placed on the surface of the human body, the movement of the human body posture will cause fluctuations in the received signal amplitude and change the transmission characteristics of the channel. Moreover, due to the high randomness and complexity of human body motion, mathematical models cannot be used to model and predict, that is, there is no general way to predict the motion of different people, so the space around the human body is defined as non-predictable free space, hereinafter referred to as non-predictable free space free space.

In the non-free space around the human body, due to the introduction of the random motion of the human body itself, the whole process cannot be described by a mathematical model, but in this space, for any infinitesimal time dt, a standard free Space, many existing linear detection methods can be used in this standard free space. After integrating the time dt to the entire time course, although the continuous integration result of the discrete motion of the human body belongs to the free space, the introduction of the attitude function will lead to irregular motion parameters and input to the final integration result, so the integration result is still a non-free space. However, since the linear detection method involved is available in the free space determined at any moment in the space, it is still applicable in the entire non-free space.

After searching the existing literature, it was found that Tevfik Yucek and others elaborated in detail in "A Survey of SpectrumSensing Algorithms for Cognitive Radio Applications (Survey of Spectrum Sensing Methods in Cognitive Radio Applications)" in IEEE Magazine, including energy detection, matched filtering There are several commonly used signal spectrum detection methods in free space, including cyclic spectrum detection and cooperative detection. According to some existing communication protocols, specific detection methods and performance analysis are proposed, and their feasibility is preliminarily demonstrated. However, this technology is developed around the wireless standard free space, and has not been specifically researched on the actual channel characteristics in the wireless body area network. Any of the independent linear detection methods can only complete signal detection in a fixed environment. Considering the complexity of human body posture changes, the method proposed in this technology cannot cope with the changing wireless environment around the human body. signal detection needs.

Contents of the invention

The object of the present invention is to overcome the above-mentioned shortcomings of the prior art, and provide a linear detection method for signals in a non-free space environment around a human body. The present invention obtains quantifiable statistical parameters that can reflect the characteristics of the received signal by using a specific preprocessing process, and applies the commonly used linear detection methods in the standard free space to the detection environment of the nonlinear space around the human body. Adjust the statistical parameters according to the characteristics of the human body, dynamically select the most suitable detection method, and avoid the limitations of any single detection method in the nonlinear space environment; at the same time, introduce the method of estimating the human body movement by detecting the human physiological signal, and follow the human body during the movement process. Motion adjusts the selected parameters and detection methods in time, thereby effectively improving the detection performance of the system.

The present invention is achieved through the following technical solutions, and the present invention comprises the following steps:

Step 1, preprocessing the received wireless signal to obtain the general parameters of the signal and the human body's electrocardiogram (ECG), and start timing.

The general parameters reflect the basic channel characteristics of the wireless transmission channel, including: signal carrier-to-interference ratio, signal carrier-to-noise ratio, RSSI (received signal strength indication) value, channel impulse response function, multipath delay distribution probability density function and The prior synchronization information of the received signal.

Step 2. According to the obtained general parameters, use the existing linear detection method to perform preliminary detection on the received signal, obtain the signal spectrum characteristics under each linear detection method, and then obtain the detection performance curve of each signal spectrum characteristic, and then detect The linear detection method with the priority of the performance curve is used as the initial detection method.

The existing linear detection methods include: matched filter method, energy detection method and cyclic spectrum detection method.

The preliminary detection is specifically: within 0.5ms-5ms, using the existing linear detection method to detect the received signal.

The linear detection method with the priority of the detection performance curve is: when the detection probability in the detection performance curve is G, the linear detection method corresponds to the smallest signal-to-noise ratio.

Step 3: Extract the human ECG signal in step 1. When the human ECG signal meets the physical realization conditions, use the initial detection method as the optimal detection method to detect the received signal; otherwise, use the energy detection method as the optimal detection method Check the received signal.

The physical realization conditions are as follows: estimate the human body motion posture on the human ECG signal, and obtain the relative position of the current node and its nearest node. When the current node and its nearest node are not blocked by the human body, the current node and its nearest node have direct contact. path, the human ECG signal meets the physical realization conditions; otherwise, the human body ECG signal does not meet the physical realization conditions.

Step 4: In the process of using the optimal detection method to detect the received signal, when the posture of the human body changes, return to Step 1 and perform detection again; otherwise, perform Step 5.

The change of the human body posture means that the error of the human body's electrocardiographic signal is greater than 5 beats of the human body's electrocardiographic signal.

Step five, when the time of the timer is less than the time threshold T, use the optimal detection method in step three to detect the wireless body area network; otherwise, return to step one and perform detection again.

The value range of the time threshold T is: 1s-10s.

Compared with the prior art, the present invention has the following beneficial effects:

1. In view of the changeable and irregular channel characteristics in the non-free space, different linear detection methods are dynamically selected. In the case of complex and changeable channels in the non-free space environment around the human body, it can effectively prevent a single detection method from being unable to cope with various The problem of channel conditions.

2. Use the available general parameters and human ECG signals in two steps, so that the human ECG signals have a higher priority, so that the system can make full use of prior information to quickly select or eliminate methods, and improve the screening efficiency of detection methods efficiency.

3. With the addition of motion estimation assistance, the selected parameters can be dynamically refreshed according to the motion of the human body, ensuring that within the preset detection cycle, sudden changes in human body posture will not have a great impact on the detection performance, which improves the detection performance around the human body. Detection performance in free space.

Detailed ways

The embodiments of the present invention are described in detail below: the present embodiment is implemented under the premise of the technical solution of the present invention, and detailed implementation and specific operation process are provided, but the protection scope of the present invention is not limited to the following implementation example.

Example

This embodiment includes the following steps:

Step 1, preprocessing the received wireless signal to obtain the general parameters of the signal and the human body ECG signal, and start timing.

The general parameters reflect the basic channel characteristics of the wireless transmission channel, including: signal carrier-to-interference ratio, signal carrier-to-noise ratio, RSSI value, channel impulse response function, multipath delay distribution probability density function and prior synchronization information of received signals .

Step 2. According to the obtained general parameters, use the existing linear detection method to perform preliminary detection on the received signal, obtain the signal spectrum characteristics under each linear detection method, and then obtain the detection performance curve of each signal spectrum characteristic, and then detect The linear detection method with the priority of the performance curve is used as the initial detection method.

The existing linear detection methods include: matched filter method, energy detection method and cyclic spectrum detection method.

The preliminary detection specifically includes: using an existing linear detection method to detect the received signal within 1 ms.

Described matched filter method is specifically: carry out A/D conversion (analog-to-digital conversion) to the received wireless signal, carry out sampling after convolution with prior synchronous signal, N sampling points are summed to obtain Decision statistic T, when T>γ, there are other nodes in the detection range; otherwise, there are no other nodes in the detection range, where: γ is the decision threshold.

The energy detection method specifically includes: performing pre-filtering on the received wireless signal, performing A/D conversion after removing out-of-band noise and other adjacent signals, and summing the conversion results to obtain the decision statistic T, when T>γ′, there are other nodes in the detection range; otherwise, there are no other nodes in the detection range, where: γ′ is the decision threshold.

The cyclic spectrum detection method specifically includes: performing A/D conversion and FFT (Fast Fourier) transformation on the received wireless signal Y[n], performing correlation calculations on the conversion results and then summing them to obtain a judgment Statistics T, when T>γ", there are other nodes in the detection range; otherwise, there are no other nodes in the detection range, where: γ" is the decision threshold.

The linear detection method with the priority of the detection performance curve is: when the detection probability in the detection performance curve is G, the linear detection method corresponds to the smallest signal-to-noise ratio.

In this embodiment, the detection performance curve obtained by the cyclic spectrum detection method is given priority, so the cyclic spectrum detection method is used as the initial detection method.

Step 3: Extract the human ECG signal in step 1. When the human ECG signal meets the physical realization conditions, use the initial detection method as the optimal detection method to detect the received signal; otherwise, use the energy detection method as the optimal detection method Check the received signal.

The physical realization conditions are as follows: estimate the human body motion posture on the human ECG signal, and obtain the relative position of the current node and its nearest node. When the current node and its nearest node are not blocked by the human body, the current node and its nearest node have direct contact. path, the human ECG signal meets the physical realization conditions; otherwise, the human body ECG signal does not meet the physical realization conditions.

In this embodiment, the current node and its closest nodes are located on both sides of the human torso, and there is no direct path, so the physical realization conditions are not satisfied. The energy detection method is selected as the optimal detection method to detect the received signal.

Step 4: During the signal detection process using the energy detection method, when the posture of the human body changes, return to Step 1 and perform detection again; otherwise, perform Step 5.

The change of the human body posture means that the error of the human body's electrocardiographic signal is greater than 5 beats of the human body's electrocardiographic signal.

Step five, when the time of the timer is less than the time threshold T, use the energy detection method to detect the wireless body area network; otherwise, return to step one, and perform detection again.

The time threshold T in this embodiment is 5s.

Through comparison, it is found that: in the channel environment where the transmitted signal energy is -35dbm, that is, a poor signal-to-noise ratio, and the false alarm probability conditions are the same, the detection probability performance obtained by using the method of this embodiment is 60% higher than that obtained by using the energy detection method. %-70%.

When the wireless environment changes rapidly, the existing single detection method is often unable to meet the actual detection needs, and the method of this embodiment can dynamically adjust the detection method according to the change of the actual detection environment to achieve the best detection performance. At the same time, the introduced motion estimation method It can effectively reduce the impact of changes in human body posture on the signal detection effect and improve the detection effect.

Claims (7)

1. a linear detection method of signal under non-free space environment around human body, is characterized in that, comprises the following steps: Step 1, preprocessing the received wireless signal to obtain the general parameters of the signal and the human body ECG signal, and start timing; Step 2. According to the obtained general parameters, use the existing linear detection method to perform preliminary detection on the received signal, obtain the signal spectrum characteristics under each linear detection method, and then obtain the detection performance curve of each signal spectrum characteristic, and then detect The linear detection method with the priority of the performance curve is used as the initial detection method; Step 3: Extract the human ECG signal in step 1. When the human ECG signal meets the physical realization conditions, use the initial detection method as the optimal detection method to detect the received signal; otherwise, use the energy detection method as the optimal detection method Detect the received signal; Step 4, during the detection process using the received signal of the optimal detection method, when the posture of the human body changes, return to step 1 and perform detection again; otherwise, perform step 5; Step five, when the time of the timer is less than the time threshold T, use the optimal detection method in step three to detect the wireless body area network; otherwise, return to step one and perform detection again. 2. the linear detection method of signal under the non-free space environment around human body according to claim 1, it is characterized in that, described common parameter reflects the basic channel characteristic of wireless transmission channel, comprises: signal carrier to interference ratio, signal carrier to noise ratio Ratio, RSSI value, channel impulse response function, multipath delay distribution probability density function and prior synchronization information of the received signal. 3. the linear detection method of signal under the non-free space environment around the human body according to claim 1, is characterized in that, the existing linear detection method described in step 2 comprises: matched filter method, energy detection method and loop Spectrum detection method. 4. The linear detection method of the signal under the non-free space environment around the human body according to claim 1, characterized in that the preliminary detection described in step 2 is: within the time of 0.5ms-5ms, using the existing The linear detection method detects the received signal. 5. the linear detection method of signal under the non-free space environment around human body according to claim 1, it is characterized in that, the linear detection method of described detection performance curve priority is: when detection probability is G in the detection performance curve, The linear detection method corresponding to the smallest signal-to-noise ratio. 6. the linear detection method of signal under the non-free space environment around the human body according to claim 1, is characterized in that, the physical realization condition described in the step 3 is: do the estimation of human body movement attitude to human body electrocardiogram signal, Get the relative position of the current node and its nearest node. When the current node and its nearest node are not blocked by the human body, there is a direct path between the current node and its nearest node, and the human body's ECG signal meets the physical realization conditions; otherwise, the human body's ECG signal does not satisfy the physical realization condition. 7. The method for linear detection of signals in a non-free space environment around the human body according to claim 1, wherein the value range of the time threshold T in step 5 is: 1s-10s.