| With the development of electronic information, military departments use radio communication more and more for command and deployment in a war. Modern war turns into a fight on the utilization and scramble of electronic resources. With various communication modes continuously generating, opposing sides concentrate on competitions of signal interception and anti-interception. As a result, intercepting and identifying means are also being improved. Radio communication is conducted in the form of electromagnetic radiation with an open space. When our own side received the sending-out signal, it is possible for belligerent to detect the signal at the same time. There is no way to prevent the invasion of enemy's jamming signal. So, it is extremely important to use a safe communication means for information transmission. In contrast, if we can find a simple and effective means of signal identification, it will play a decisive role in the outcome of modern war. In this background, researches on non-cooperative burst signal demodulation will bring great application value and practical significance. And this kind of demodulation is only based on sparse decomposition.Burst communication transmitter just does a short-time work. During most of the time, it keeps silent. To a certain extent, this operating state not only can avoid enemy's interference and direction to reduce the interception opportunities, lower the requirement on transmitter's power, but also can sharply cut its volume. Besides, it can shorten the occupied time of signal channel. Relative to the normal signal, burst signal's presentation time is random and the duration is very short. It possesses strong confidence, anti-interference, and other properties. In the situation of non-cooperative receiving, it is difficult to demodulate a burst signal, due to the shortage of necessary information. Therefore, the burst signal became one kind of communication signals which commonly used in military information resistance.As a new method in signal decomposition, the signal sparse decomposition has many good features. It reflected frequency, amplitude and phase on the sparse components through the atomic decomposition and avoided the carrier synchronization. It has been widely applied in signal processing, such as signal compression coding, signal identification, signal time-frequency distribution research and so on. The non-cooperative burst communication signal is the principal research object in this paper; we take the sparse decomposition theory into signal demodulation process, and propose a new paradigm in non-cooperative burst signal parameter estimation and demodulation.First of all, we introduced the establishment process between two communication sides using burst signal, also introduced some interception measures of non-cooperative signal. Simultaneously, we explained the basic principles and flows of modulation and demodulation with burst communication signal, then pointed out that the present demodulation systems is powerless on burst signal demodulation, especially on demodulation of non-cooperative receiving's. Compared with advantages and disadvantages of the commonly used methods in signal indication field, and also considering the sparse characteristic on certain transformation territory performed by communication signal, we suggest that a brand-new signal theory can be applied into burst signal demodulation. In the next two chapters, we begin with the sparse decomposition theory discussion from optimization algorithm for designing and establishment of over-complete dictionaries, and made corresponding improvement and optimization. This can reduce computation and operation complexity, and obtain a great improvement on the whole. Then, we took improved sparse decomposition theory into signal de-noising and parameter estimation, and focused on some critical steps and key issues. Eventually, using comparable simulations on views and opinions the paper involved, we verified the theoretical study's feasibility and reliability. |
PDF Full Text Request