The thesis illustrates a DSP-based speaker recognition system . In order to make the modular within the representation floating-point, we simplify the algorithm. This speaker recognition system is including hardware setting and implementation of speaker algorithm. The DSP chip is float arithmetic DSP(ADSP-21161 of ADI SHARK Series) , the algorithm of speaker recognition is gaussian mixture model. According to result of experiments, the speaker recognition of DSP can gain good recognition and speed efficiency.

In this thesis, the PC base and DSP base speech keyword retrieval and recognition systems could work. The keywords and describing sentences will not have the limit of word length and could be any languages. Besides, training speech models is not needed anymore. It means that the database gets its expansibility without training speech models again. We can establish the system on the PC base, and calculate the program with fixed-point DSP board. In the processing of speech signal, lots of mathematical functions will be required. We must reach its immediately effect, so that the system could be useful. In addition, compared with floating point, the fixed point DSP cost much less; it makes the system nearer to users. After being tested by experiments, the speech keyword retrieval and recognition system got great recognition and efficiency.

The main goal of this thesis is to design and implement the high performance processor core for completing those digital signal processing algorithms applied at the DVB-T systems. The DSP must support the signal flow in time. Completing the FFT algorithm at 8192 input signal points instantaneously is the most important key. In order to achieve the time demand of FFT and the DSP frequency must be as lower as possible, the way is to increase the degree of instruction level parallelism (ILP). The thesis designs a VLIW architecture processing core called DVB-T DSP to support instruction parallelism with enough execution units. The thesis also uses the software pipelining to schedule the loop to achieve the highest ILP when used to execute FFT butterfly operations. Furthermore, in order to provide the smooth data stream for pipeline, the thesis designs a mechanism to improve the modulo addressing, called extended modulo addressing, will collect the discrete vectors into one continuous vector. This is a big problem that the program size is bigger than other processor architecture at the VLIW processor architecture. In order to solve the problem, this thesis proposes an instruction compression mechanism, which can increase double program density and does not affect the processor execution efficiency. The simulation result shows that DVB-T DSP can achieve the time demand of FFT at 133Mhz. DVB-T DSP also has good performance for other digital signal processing algorithms.

This thesis mainly focuses on active power factor correction (PFC) circuit research. Based on DSP, a circuit with a digital controller is developed. Boost converter is used as the power stage. For power factor correction, the control law is realized with DSP and the control strategy of controller adopts the multi-loop method, i.e. voltage-loop and modified PI current-loop. Finally, compared with the circuit using the conventional PI control law, performances of the modified PI current-loop circuit are obviously optimized.

<p>Uppgiften består i att konstruera och designa en DSP lösning för att förebygga stamning. Genom att en signal tas in på en mikrofon och bearbetas i en DSP därefter skickas ut på en hörlur.</p><p>Signalen som behandlas i DSP:n ska frekvenshöjas och fördröjas.</p><p>Pitch och fördröjning ska kunna justeras m.h.a. en omkopplare.</p><p>Till hjälp fanns utvecklingskortet eZdspF2808 från Texas Instruments och en in/utgångsförstärkare.</p><p>Tidigare erfarenheter av DSP från Texas Instruments saknades. Att lära känna utvecklingskortet var därför en del av arbetet. Det bestod mest i att läsa användarguider från Texas Instruments hemsida.</p><p>Utvecklingskortet saknar en DA-omvandlare, en sådan konstruerades.</p> / <p>This thesis describes a system that uses DSP technology to reduce stuttering. One eZdspF2808 development card is used as a test platform. The lack of a DA-converter forced us to construct our own.</p><p>A software algorithm producing a delay and a pitch was implemented in DSP C programming language.</p>

Στη παρούσα εργασία περιγράφεται η σχεδίαση και η υλοποίηση μιας θέσεως εργασίας (workstation) ενός εργαστηρίου DSPs από απόσταση που θα εξειδικεύεται στην επεξεργασία εικόνας. / This master thesis describes the design and implementation of a Workstation of a DSPs Remote Lab, specialized in image processing.

In the final work it is made analysis of Petri nets and Petri nets simulators, analysis of modeling system composition and its methodology. There were presented topicality of theme, it’s practical and theoretical meaning, work purpose and tasks. There was made analysis of Petri net simulator on DSP and was made realization of Petri net on DSP. Presented conclusions of the final work, therefore it was mentioned practical results.

Wireless sensor network technology has now been widely adopted. In many applications, distributed sensor nodes collect data at different locations and the location information of each node is required. The Global Positioning System is commonly used to identify the location of the nodes in such networks. Although GPS localization has consistent long-term accuracy, it is limited by the inherent dependency on a direct line of sight to 4 or more external satellites. The increasing demand for an embedded system providing reliable navigation solutions regardless of its operational environment has motivated investigations into the use of integrated systems that combine inertial sensors with GPS receivers. This research proposes a hardware architecture for location-based wireless sensor networks. In this architecture, each sensor node consists of a GPS receiver, a reduced set of low cost micro-electro-mechanical-system-based INS and a wireless transceiver. Sensor nodes in WSN are often equipped with irreplaceable batteries, which makes the power consumption crucial. To reduce the energy consumption, a microcontroller is used to control the power supply. Besides, a motion detection scheme is proposed by taking advantage of the ultra low-power wake-up function of the microcontroller. A low-power featured digital signal processor is used to accomplish the navigation computation using the Kalman filter for GPS/INS data fusion. Non-Holonomic Constraints derived velocity updates are applied to reduce the position errors. Field tests are conducted to verify the real-time performance of the proposed system with a positioning update rate of 20 Hz. The first test shows that the 2D INS/GPS integration can maintain the average system position error within 5 meters during a 60-second GPS outage. The second test used low cost inertial sensors. The average position error was 10.17 meters during a 20-second outage. The largest RMS value of position errors among these outages was within 14.5 meters. Furthermore, additional accuracy improvements of approximately 1.4 meters were achieved by utilizing NHC during GPS outages. The third test shows that the average error during a 30-second outage is approximately 20.6 meters for the on-foot scenario and 26.7 meters for the in-vehicle scenario. / Thesis (Master, Electrical & Computer Engineering) -- Queen's University, 2012-01-13 14:46:45.44

Music today is a big part of a person's everyday life and can influence emotions as wellas initiative and motivation. This effect together with development in technology hasled to that music today is accessible and can be played from a large amount of differentkinds of products. This in turn has meant that the demand for smaller audio playbackdevices is greater than ever.The goal of this project is to develop a concept for energy-efficient speakers withwireless communication that has the ability to obtain and play high quality music. Thespeakers will use the latest available technology and at the same time, using digitalsignal processing, have a low environmental impact through high energy efficiencyand low power consumption. Parallel to this work a pilot study will be conducted tosee trends in listener preference of sound.The result achieves the goals pledged while power consumption in sleep mode is solow that they meet the demands of tomorrow. It also shows that if the time and thewill, and if financial resources are available, it is possible to combine the reducedenvironmental impact with sound quality even for the discerning listener. / Musik är idag en stor del av en persons vardag och kan påverka såväl känslor sominitiativ och motivation. Denna inverkan tillsammans med teknikutvecklingen har letttill att musik idag finns tillgängligt och kan spelas från en stor mängd olika sortersprodukter. Detta har medfört att efterfrågan på mindre ljuduppspelningsenheter ärstörre än någonsin.Målet med detta projekt är att ta fram ett koncept på en energieffektiv högtalare medtrådlös kommunikation som har förmågan att med hög kvalité spela upp musik, användasenaste tillgängliga teknik och samtidigt, med hjälp av digital signalhantering, ha lågmiljöpåverkan genom lägre effektförluster och låg strömförbrukning. Parallellt meddetta kommer en förstudie genomföras för att se trender i lyssnares önskemål avljudkaraktär.Resultatet uppnår ställda mål samtidigt som effektförbrukningen i viloläge är så låg attden möter morgondagens krav. Detta visar att om tid, vilja och ekonomiska resurserfinns att tillgå så är det möjligt att förena minskad miljöpåverkan med ljudkvalité ävenför den kräsne lyssnaren.

Telemedicine systems represent significant achievements in the provision of clinical medicine and health care service using telecommunication and information technologies for the purpose of remote monitoring. Almost all telemedicine systems require a network-enabled device, ranging from server machine to smart phone, which delivers the data as a transmission gateway. The research in this thesis introduces the hardware and software design of a novel Internet-based real time DSP system in the application of telemedicine. Before this work, it was not previously achievable or economically feasible to develop a telemedicine system with a truly embedded measurement platform for real-time monitoring of clinical information on a global scale. The novelty of this design consists in embedding the Internet-based monitoring into the real-time signal processing system, as well as incorporating the merits of wireless communication and global distributed measurement. To demonstrate this concept, a prototype of a truly embedded device incorporating either a browser-based application or a LabVIEW software application has been designed and developed, which is able to provide real-time biomedical signal acquisition, processing, wired/wireless transmission, visualisation, storage and retrieval via the Internet. The concept-to-prototype manipulates multiple biomedical signals from multi-sensors during studies and distributes them to the Internet. The prototype was evaluated on volunteers in vivo under ethical approval. The designed system was also tested under various physiological conditions and different Internet speeds. It manifests desirable performance regarding multi-functionality, ubiquitous accessibility, robustness, and adaptability. The full functionality of this innovative system successfully enables clinicians to remotely monitor a patient's physical condition in real-time globally. The experimental results obtained from the host are in close agreement with the expected performance of the designed system, which proffers evidence that this system represents a true innovation in the realm of telemedicine.