Font Size: a A A

Research On And Applications Of Adaptive Signal Detection Techniques For High Frequency Channels

Posted on:2000-02-09Degree:DoctorType:Dissertation
Country:ChinaCandidate:H ZhangFull Text:PDF
GTID:1118359972950020Subject:Communication and Information System
Abstract/Summary:PDF Full Text Request
uswnFor a 1ong time, short--wave communication has been an importantconununication means because of its inherent characterist ics, especial1yin military fie1d, and is an indispensable component of our mi1itaryconununication at all times. High--speed data transmission in a short--wavechannel is an important research topic in the fie1ds of communicationsand electronic systems.This paper 1ays emphasis on the author's work, some new ideas and newachievements in the research fields of adaptive signal detectiontechniques, short--wave modem's design and its realization. First,according to the characteristics of the short--wave channels, itsinfluence on signal transmission is analyzed, and in the light of theM--phase modulation, both the equivalent baseband model and the equivalentvoiceband model of the short--wave channels are provided.Second1y, on the basis of the characteristic that the output from theshort--wave channel can be equivalent to the convolution integral of theinput signa1 and the unit impulse response, the convolutionalhomomorphism filter is constructed, and some problems inc1uding theadaptive channel estimation, the estimtion interpolat ing comb fi ltering,the complex logarithIn numerical analysis the phase extension and so onare studied so that the influence of multipath signals is decreased, thequa1ity of output signals is improved, and a valuable conclusion isobtained. Based on the construction of the decision feedback equa1izer(DFE), the decision feedback equa1ization technique re1ated to thefractional interval and the hybrid--algorithm combined estimation and themethod with which controlling the tapping coefficients of an equa1izerare presented, thus simp1ifying the construction of the system, reducingthe amount of operation, preventing the equalizer from diffusing,enhancing the capabi lity of the equal izer to prevent burst interferenceand deep fading, and also playing an imPortant role in high--speed modem'sresearch and design. By the MLSE theory, an NMLSE scheme is provided. ManysiIIlP1ified methods and a channel evaluator which uses a variable step LMSa1gorithIn and a prognostic algorithIn are researched, so as to makeevaluated values of the channel?s status more reasonable, simplify the system?s construction, and reduce the operation amount, which is of important academic and practical significance. On the basis of the least sum of squared errors(LSSE) criterion, SBE is discussed, in which the real梩ime tracing of the channel parameters is realized and the data is estimated with the known training symbols. By adopting the LMS algorithm, the impulse response of the HF channels is evaluated, and then a matrix equation of the channel?s parameter is obtained, so that the restrictive conditions of DDE are canceled and the length of both the training block and the source data block can be reduced. A promising adaptive signal detection method is to adopt the form with the ruling decision feedback and simplified Cholesky matrix disintegrated algorithm, which can reduce the complexity, increase the speed of data transmission and throughput efficiency.Finally, Prototypes which can accomplish the functions of a modem with the short梬ave serial system are realized. In the prototypes, a high梥peed digital signal processor is utilized as the core, and the serial system techniques and many interference梖ree techniques are adopted. Some laboratory tests and routine tests in the short梬ave channels on the prototypes prove that the prototypes satisfy the requirement.
Keywords/Search Tags:Short-wave communication, Data transmission, Modem, Adaptive equalization, Decision feedback equalizer, Near-maximum-likelihood estimation technique, Data directed estimation technique, Block data detection technique, Homomorphism filtering technique
PDF Full Text Request
Related items