• Proceedings of the IEEK Conference
• /
• summer
• /
• pp.391-394
• /
• 2004

This paper focuses on the DSP implementation of an HMM-based speech recognizer that can handle several hundred words of vocabulary size as well as speaker independency. First, we develop an HMM-based speech recognition system on the PC that operates on the frame basis with parallel processing of feature extraction and Viterbi decoding to make the processing delay as small as possible. Many techniques such as linear discriminant analysis, state-based Gaussian selection, and phonetic tied mixture model are employed for reduction of computational burden and memory size. The system is then properly optimized and compiled on the TMS320C6711 DSP for real-time operation. The implemented system uses 486kbytes of memory for data and acoustic models, and 24.5kbytes for program code. Maximum required time of 29.2ms for processing a frame of 32ms of speech validates real-time operation of the implemented system.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.29 no.12
• /
• pp.83-87
• /
• 2015

This paper presents a simple notch filter, which is so suitable for three-phase unbalanced and distorted power line. In the d-q synchronous transformation, three-phase unbalanced and distorted voltages generate lots of ripple voltages on d-q axes. The ripples make disturbances on controllers such as PLL of phase tracking. Unbalanced state makes ripple of double the frequency of power line. Odd harmonics 5th and 7th on the line make even 4th and 6th ripples on d-q axes due to the rotating reference frame, respectively. Cascaded two comb filters, delay lines 1/4T and 1/8T, are adopted for the ripple rejection. The filter rejects harmonics 2nd, 4th, 6th, 10th and so on. They are very effective to remove the ripples of both unbalance and distortion. The filter, implemented by two FIFOs on an experimental system, is adopted on a PLL controller of power line phase tracking. Through the simulation and experimental results, performance of the proposed comb filter has been validated.

• The KIPS Transactions:PartC
• /
• v.9C no.6
• /
• pp.953-960
• /
• 2002

Rel-time application programs have constraints which need to be met between media-data. It is client-leading synchronization that is absorbing variable transmission delay time and that is synchronizing by feedback control and palyout control. It is the important factor for playback rate and QoS if the buffer level is normal or not. This paper, The method of maintenance buffer normal state transmits in multimedia server by appling feedback of filtering function. And synchronization method is processing adaptive playout time for smooth presentation without cut-off while media frame is skip. When audio frame which is master media is in upper threshold buffer level we decrease play out time gradually, low threshold buffer level increase it slowly.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.05a
• /
• pp.281-286
• /
• 2009

The proposed technique uses cyclic frame structure, where three periods such as beacon period (BP), mesh contention access period (MCAP) and slotted period (SP) are in a data frame. This paper studies on a mechanism to allow communication nodes (6lowpan) in a PAN with different logical channel for global healthcare applications monitoring technology. The proposed super framework structure system has installed 6lowpan sensor nodes to communicate with each other. The basic idea is to time share logical channels to perform 6lowpan sensor node. The concept of 6lowpan sensor node and various biomedical sensors fixed on the patient BAN (Body Area Network) for monitoring health condition. In PAN (hospital area), has fixed gateways that received biomedical data from 6lowpan (patient). Each 6lowpan sensor node (patient) has IP-addresses that would be directly connected to the internet. With the help of IP-address service provider can recognize or analyze patient data from all over the globe by the internet service provider, with specific equipments i.e. cell phone, PDA, note book. The NS-2.33 result shows the performance of data transmission delay and data delivery ratio in the case of hop count in a PAN (Personal Area Networks).

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.1
• /
• pp.61-67
• /
• 2022

In this paper, we propose a model that can convert some of the simulation program resources to a mobile environment. Recently, smart factories that use PLCs as controllers in the manufacturing industry are rapidly becoming widespread. However, in the situation where it is difficult to operate due to the shortage of PLC operation personnel, the actual situation is that a platform for PLC operation education is necessary. Currently most PLC-related educational platforms are based on 2D, which makes accurate learning difficult and difficult. When a simulation program is applied to distance learning in a general PC environment, many elements are displayed on the monitor, which makes screen switching inconvenient. Experiments with the proposed model confirmed that there was no frame deterioration under general circumstances. The average response time by the request frame was 102 ms, and it was judged that the learner was not at the level of experiencing the system delay.

• The Journal of the Acoustical Society of Korea
• /
• v.25 no.8
• /
• pp.383-388
• /
• 2006

This Paper presents a novel approach to single channel microphone speech enhancement in noisy environments. Widely used noise reduction techniques based on the spectral subtraction are generally expressed as a spectral gam depending on the signal-to-noise ratio (SNR). The well-known decision-directed(DD) estimator of Ephraim and Malah efficiently reduces musical noise under the background noise conditions, but generates the delay of the a prioiri SNR because the DD weights the speech spectrum component of the Previous frame in the speech signal. Therefore, the noise suppression gain which is affected by the delay of the a priori SNR, which is estimated by the DD matches the previous frame rather than the current one, so after noise suppression. this degrades the noise reduction performance during speech transient periods. We propose a computationally simple but effective speech enhancement technique based on the sigmoid type function for the weight Parameter of the DD. The proposed approach solves the delay problem about the main parameter, the a priori SNR of the DD while maintaining the benefits of the DD. Performances of the proposed enhancement algorithm are evaluated by ITU-T p.862 Perceptual Evaluation of Speech duality (PESQ). the Mean Opinion Score (MOS) and the speech spectrogram under various noise environments and yields better results compared with the fixed weight parameter of the DD.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.9A
• /
• pp.1285-1294
• /
• 1999

In this paper we consider the performance of TCP video traffic over ABR with Long-Range Dependent VBR traffic. As compressed coded video traffics are increasing rapidly over Internet, lots of studies are being done for transmitting those traffics efficiently using limited network resources. We consider here the transmitting video service over ABR service in ATM networks, especially satellite networks. CBR or VBR services are suggested in transmitting the video traffic in ATM Forum TM 4.0. But ABR service connection, which is considered as appropriate service for data traffic, can be established with a small amount of bandwidth, MCR (Minimum cell rate). Furthermore ABR service can control the source's transmitting rate using feedback mechanism. Using this feature ABR service can be used in some applications which can control their quality of services corresponding to network loads. Compressed video sources with MPEG-2 are used for Long-Range Dependent VBR traffic here. We model the compressed video source to resemble the MPEG-2 transport streams. These compressed video traffic streams are consisted of three different frames, I-frame, P-frame, and B-frame. So when a network are overloaded, we can control the quality of service using this traffic features. TCP Traffics over ABR need large buffers in ATM switch to satisfy their QoS with background VBR traffics, which have high deviations in bandwidth. Furthermore satellite ATM networks with large feedback delay need large buffers corresponding RTT delay. The performance comparisons among EFCI and ER switch (ERICA+) switches in the network circumstances described above were shown in this paper. We also considered the case with ON-OFF VBR traffics.

• Journal of IKEEE
• /
• v.7 no.2 s.13
• /
• pp.288-298
• /
• 2003

We consider a window-controlled selective-repeat ARQ scheme for error control between two adjacent nodes lying on a communication path. In this scheme, each packet to be transmitted is endowed with a serial number in a cyclic and sequential fashion. In turn, the transmitting node is not allowed to transmit a packet belonging to a window before every packet in the previous window is positively acknowledged. Such postponement of packet transmission incurs a degradation in throughput and delay performance. In this paper, aiming at improving packet delay performance, we employs a supplement scheme in which a serial number is duplicated within a frame. Classifying duplication rules into fixed, random and adaptive categories, we present candidate rules in each category and evaluate the packet delay performance induced by each duplication rule. From numerical examples, we observe that duplicating serial numbers, especially ADR-T2 effectively reduces mean packet delay for the forward channel characterized by a low packet error rate. We also reveal that such delay enhancement is achieved by a high probability of hitting local optimal window size.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.2B
• /
• pp.117-128
• /
• 2006

For the legacy IEEE 802.15.3 WPAN protocol, an unpredictable piconet coordinator(PNC) leaving from a piconet without a proper handoff procedure causes an absence of PNC, and thus the piconet gets collapsed. In addition, several beacons from PNCs in adjacent piconets may be collided on a device(DEV) located between those piconets. This beacon collision eventually makes the DEV leave from the piconet. To remedy these two problems, we here propose an Active Seamless Coordinator Switching(ASCS) scheme and a PNC HB based Beacon Alignment(PHBA) one. In the ASCS scheme, a PNC assigns a number of DEVs as next possible PNCs in sequence for provisioning against the abrupt breakdown of the current active PNC. Each nominated DEV proactively sends a probe frame to confirm the operation status of the active PNC. For the case of no response from the PNC, the nominated DEV tries to become a new PNC immediately. In the second PHBA scheme, each PNC is allow to broadcast a special Heart Beat(HB) frame randomly during a superframe period. When a DEV receives a HB frame from other PNC, it promptly sends the related PNCs a special Hiccup Beat(HCB) frame with the superframe information of its associated PNC. As a result, the HCB frame makes both PNCs align their superframe beginning time in order to yield no more beacon collisions. For these two proposed schemes, we show the performance by simulations. We can confirm the enhancement of throughput for each superframe and average frame transfer delay, since each scheme can reduce the duration of piconet collapse. Finally, it is worth while to note that the proposed schemes can be operated with frames those are permitted in the legacy WPAN standard.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.43 no.2 s.344
• /
• pp.56-65
• /
• 2006

In this paper, we introduce modified difference type symbol timing detection method of simple structure and show the relations between S/N ratio and timing detection performance which less influenced by multipath channel delay profile and added noise level and it show very exact GI detection performance characteristics. In the computer simulations, 4 symbol time duration of short and long training of IEEE802.11a standard OFDM frame are used for symbol synchronization timing detection. The computer simulation results show the very exact symbol timing detection performance characteristic within 1 sample error of OFDM signal regardless channel delay profile from minimn phase channels of phase rotation ${\pi}/2$ to non-minimum phase channels of phase rotation ${\pi}/2$ of received OFDM signal and added noise level in channel.