Filter by type:

Sort by year:

1- Distributed Diffusion Based Spectrum Sensing for Cognitive Radio Sensor Networks Considering Link Failure

Journal paper
Amirhosein Hajihoseini Gazestani and Seyed Ali Ghorashi


Wireless sensor networks (WSNs) typically use license free industrial, scientific and medical (ISM) frequency bands and by increasing the demand for different usages of WSNs, it is anticipated that this band gets saturated. One solution to deal with this spectrum scarcity in WSNs is utilizing the cognitive radio concept, called cognitive radio sensor networks (CRSNs). One of the most critical challenges in CRSNs is spectrum sensing, which can be implemented by centralized, cluster based or distributed methods. In CRSNs, distributed methods have better spectrum sensing performance due to their fast adaptation to the network changes. Also they have lower power consumption level, which is critical in CRSNs. In this paper, we propose a novel distributed diffusion based spectrum sensing method for CRSNs that improves the robustness of the spectrum sensing method against link failure and network topology changes. It also increases the algorithm’s convergence rate while its accuracy is acceptable. We prove the proposed method convergence and calculate its mean square error, considering link failure assumption. Simulation results confirm that the proposed method improves the convergence rate compared with conventional distributed methods, and using the proposed method increases the convergence rate and needs less communications to make a decision about absence or presence of the primary user.
Distributed, Cooperative Communications, Consensus, Diffusion, Spectrum Sensing, Cognitive Radio Sensor Networks (CRSNs), Link Failure.

2- Introduction and Patent Analysis of Signal Processing for Big Data

Book Chapter
Mohammad Eslami, Amirhosein Hajihoseini Gazestani and Seyed Ali Ghorashi


Big data is rapidly considered in different scientific domains, industries and business methods. Considering the concept of Internet of Things, big data is generated by everything around us continuously, and therefore, dealing with big data and its challenges are important and requires new thinking strategies and also techniques. Signal processing is one of the solutions that is utilized with big data in most scientific fields. This paper gives a brief introductory preview of the subjects included in this area and describes some of challenges and tactics. In order to show the rapid growth of attentions in signal processing for big data, a statistical analysis on corresponding patents is considered as well.

3- Image dataset for Persian Road Surface Markings

Conference paper
Seyed Hamid Safavi, Mohammad Eslami, Aliasghar Sharifi, Amirhosein Hajihoseini, Mohammadreza Riahi, Maryam Rekabi, Sadaf Sarrafan, Rahman Zarnoosheh, Ehsan Khodapanah Aghdam, Sahar Barzegari Banadkoki, Seyed Mohammad Seyedin Navadeh, Farah Torkamani-Azar

Self-driving and autonomous cars are hot emerging technologies which can provide enormous impact in the near future. Since an important component of autonomous cars is vision processing, the increasing interest for self-driving cars has motivated researchers to collect different relative image datasets. Hence, we collect a comprehensive dataset about the road surface markings which are available in Iran. In addition, we evaluate the conventional recognition rate. In this paper, we present a novel and extensive dataset for Persian Road Surface Markings (PRSM) with ground truth labels. We also hope that it will be useful as a Persian benchmark dataset for researchers in this field. The dataset consists of over 68,000 labeled images of road markings in 18 popular classes. It also contains road surface markings under various daylight conditions. Our dataset with further details is available online at:

4- A Fingerprint method for Indoor Localization using Autoencoder based Deep Extreme Learning Machine

Journal paper
Zahra Ezzati Khatab, Amirhosein Hajihoseini, Seyed Ali Ghorashi


By growing the demand for location based services in indoor environments in recent years, fingerprint based indoor localization has attracted many researchers’ interest. The fingerprint localization method works based on received signal strength (RSS) in Wireless Sensor Networks (WSNs). This methods uses RSS measurements from available transmitter sensors, which are collected by a smart phone with internal sensors. In this paper, we propose a novel algorithm that takes the advantages of deep learning, extreme learning machine (ELM) and high level extracted features by autoencoder, to improve the localization performance in the feature extraction and the classification. Furthermore, as the fingerprint database needs to be updated (due to the dynamic nature of environment), we also increase the number of training data, in order to improve the localization performance, gradually. Simulation results indicate that the proposed method provides a significant improvement in localization performance, by using high level extracted features by autoencoder, and increasing the number of training data.


Indoor localization, fingerprint, wireless sensor network, autoencoder, deep extreme learning machine

5- Distributed Spectrum Sensing for Cognitive Radio Sensor Networks using Diffusion Adaptation

Journal paper
Amirhosein Hajihoseini and Seyed Ali Ghorashi


Cognitive radio is a practical solution for spectrum scarcity. In cognitive networks, unlicensed (secondary) users should sense the spectrum before any usage, to make sure that the licensed (primary) users do not use the spectrum at that time. Due to the importance of spectrum sensing in cognitive networks, this should be fast and reliable, particularly in networks with communication link failure, which leads to network topology change. Decentralized decision making algorithms are known as a promising technique to provide reliability, scalability and adaptation, especially in sensor networks. In this paper, we propose a distributed diffusion based method in which, secondary users (sensors) cooperate to improve the performance of spectrum sensing. The proposed method provides a significant improvement in convergence rate and reliability. Simulation results indicate that the proposed algorithm shows an acceptable performance and converges twice faster than recently proposed consensus based spectrum sensing algorithms in the literature, and is almost insensitive to communication link failure.


Distributed, Consensus, Diffusion, Spectrum Sensing, Cognitive Radio Sensor Networks

6- Decentralized Consensus Based Target Localization in Wireless Sensor Networks

Journal paper
Amirhosein Hajihoseini, Reza Shahbazian and Seyed Ali Ghorashi


Target localization is an attractive subject for modern systems that utilize different types of distributed sensors for location based services such as navigation, public transport, retail services and so on. Target localization could be performed in both centralized and decentralized manner. Due to drawbacks of centralized systems such as security and reliability issues, decentralized systems are become more desirable. In this paper, we introduce a new decentralized and cooperative target localization algorithm for wireless sensor networks. In cooperative consensus based localization, each sensor knows its own location and estimates the targets position using the ranging techniques such as received signal strength. Then, all nodes cooperate with their neighbours and share their information to reach a consensus on targets location. In our proposed algorithm, we weight the received information of neighbour nodes according to their estimated distance toward the target node. Simulation results confirm that our proposed algorithm is faster, less sensitive to targets location and improves the localization accuracy by 85% in comparison with distributed Gauss–Newton algorithm.


Wireless sensor network, Localization, Consensus, Weighted, Cooperative. 

7- موقعیت ‏یابی سه‏ بعدی درون ساختمانی با استفاده از سامانه ‏های مخابرات نور مرئی

Conference paper
امیرحسین حاجی حسینی، اکبر درگاهی، سیدعلی قرشی


با توجه به افزایش استفاده از طیف فرکانسی و محدودیت ‏های طیفی، انتظار می‏رود مخابرات نوری به طور گسترده در سامانه ‏های جدید استفاده شود. مخابرات نور مرئی یکی از شاخه‏ هایی است که در سالیان اخیر مورد توجه قرار گرفته و یکی از مهم­ترین کاربردهای آن موقعیت ‏یابی درون ساختمانی است، زیرا درون ساختمان نمی‏توان به خوبی از سامانه موقعیت ‏یابی جهانی استفاده کرد و استفاده از سامانه ‏های نوری به عنوان جایگزین روش مناسبی است. در این مقاله یک روش برای موقعیت‏ یابی درون ساختمانی ارائه می‏شود که فقط به اتصال گیرنده و فرستنده ‏های نوری وابسته است. در این روش پیشنهادی با استفاده از پهنای میدان نوری، زاویه سیگنال دریافتی در گیرنده تخمین ‏زده می ­شود و با استفاده از تخمین ‏زننده حداقل مربعات، موقعیت گیرنده تعیین می ‏شود. نتایج شبیه ‏سازی­ ها نشان می ‏دهد که با استفاده از چهار دستگاه دسترسی نوری می‏ توان موقعیت یک گیرنده مخابرات نور مرئی را در یک فضای سه بعدی با خطایی در حدود 0.6 متر تخمین زد.

واژگان كليدي: موقعیت‏ یابی، مخابرات نور مرئی، تخمین‏ زننده‏ حداقل مربعات، مخابرات نوری.

8- Distributed Target Localization in Wireless Sensor Networks using Diffusion Adaptation

Journal paper
Amirhosein Hajihoseini, Seyed Ali Ghorashi


Localization is an important issue for wireless sensor networks. Target localization has attracted many researchers who work on location based services such as navigation, public transportation and so on. Localization algorithms may be performed in a centralized or distributed manner. In this paper we apply diffusion strategy to the Gauss Newton method and introduce a new distributed diffusion based target localization algorithm for wireless sensor networks. In our proposed method, each node knows its own location and estimates the location of target using received signal strength. Then, all nodes cooperate with their neighbors and share their measurements to improve the accuracy of their decisions. In our proposed diffusion based algorithm, each node can localize target individually using its own and neighbor’s measurements, therefore, the power consumption decreases. Simulation results confirm that our proposed method improves the accuracy of target localization compared with alternative distributed consensus based target localization algorithms.  Our proposed algorithm is also shown that is robust against network topology and is insensitive to uncertainty of sensor nodes’ location.

Keywords: distributed; cooperative; target localization; wireless sensor network; Gauss Newton

9- مروری بر روش های موقعیت یابی در شبکه های حسگر بی سیم با استفاده ازالگوریتم های شناسایی آماری الگو

Conference paper
امیرحسین حاجی حسینی، رضا شهبازیان، سید علی قرشی، داوود غرویان


استقرار یک سامانه ­ی موقعیت ­یابی برای گره ­های حسگر، یک مسئله ­ی اساسی برای بسیاری از کاربردها در شبکه ­های حسگر بی­ سیم است. از آنجا که ممکن است شبکه­ ی حسگر در منطقه­ ای غیرقابل دسترس مستقر شده باشد، ممکن است موقعیت حسگرها از قبل مشخص نباشد. بنابراین یک سامانه ­ی موقعیت­ یابی لازم است تا اطلاعات مکانی و موقعیت گره ­ها را در اختیار قرار دهد. در بسیاری از موارد قیمت و منابع محدود انرژی، اجازه ‏ي تجهیز تمامي حسگر­ها را به سامانه ‏ي موقعیت ­یابی جهانی نمی ­دهد. لذا روش­ هایی لازم است تا گره ­ها و حسگرها بتوانند موقعیت خود را در شبکه تخمین زنند. در این مقاله سعی داریم روش­ های مبتنی بر الگوریتم ­های شناسایی آماری الگو که در زمینه ­ی موقعیت ­یابی در شبکه ­ی حسگر بی ­سیم استفاده شده ­اند را معرفی و عملکرد آن­ها را مقایسه نماییم. با توجه به کاربردهای گسترده ­ای که الگوریتم­های شناسایی آماری الگو دارند، این مبحث در سال ­های اخیر توجه زیادی را در شاخه ­های مختلف علوم به خود جلب کرده است و می­تواند برای دسته ‏بندی اطلاعات و شناسایی الگوی آن ‏ها به کار رود.

کلمات کلیدی:

شبکه­های حسگر بی­سیم، موقعیت­ یابی، شناسایی آماری الگو، KNN، ماشین بردار پشتیبان، مدل مخفی مارکف.

این مقاله آبان 1394 در هجدهمین کنفرانس ملی دانشجویی مهندسی برق در مشهد ارائه و به عنوان مقاله برتر مخابرات سیستم انتخاب شد.