Sualtı Duyarga Ağları (SDA) insan eliyle tehlikeli veya olanaksız sayılabilecek sualtı görevlerinde kullanılırlar. SDA’lar okyanusbiliminde, deprem ve tsunami tahmininde, askeri uygulamalarda, okyanus petrol platformlarının gözetlenmesinde ve çeşitli birçok alanda kullanılabilirler. Sualtı duyarga düğümleri sabit bir platforma tutturulmuş olabilir veya su içersinde serbest halde yüzebilir. Su yüzeyinden metrelerce aşağıda yüzen duyarga düğümler, gezgin sualtı duyarga ağını oluştururlar. Duyarga düğümler okyanuslardan sıcaklık, akıntı hızı, tuzluluk ve görüntü kaydı gibi veriler toplarlar. Gezgin bir SDA’da, duyarga düğümler su yüzeyinin altında, akıntıyla birlikte hareket eder ve belirli bir olayı izlerler. SDA’larda en ciddi sorunlardan biri konumlandırmadır. Konum bilgisine, veri etiketleme ve konum-tabanlı yönlendirme protokollerinde ihtiyaç duyulur. Geniş öl-çekli, üç boyutlu SDA’lar için, literatürde az sayıda konumlandırma protokolü önerilmiştir. Bu ma-kalede, İner-Çıkar düğümlerle Konumlandırma (İÇK) ve Vekil Konumlandırma (VK) yöntemlerini tanıtıp, sözkonusu yöntemlerin başarımını önceden önerilmiş olan bir başka yöntemle karşılaştır-maktayız. Bu yöntem, Geniş Ölçekli Konulmandırmadır (GÖK). Bu üç tekniğin avantaj ve dezavan-tajlarını gezgin bir SDA için göstermekteyiz. Benzetim sonuçlarımız GÖK’ün yüksek konumlandır-ma başarısına sahip olduğunu, ancak bu yöntemin beraberinde yüksek enerji tüketimi ve ek haber-leşme yükü getirdiğini göstermektedir. İÇK ise, yüksek konumlandırma başarımı, yüksek kesinlik, düşük enerji tüketimi ve düşük haberleşme maliyetine sahiptir. VK ise, kabul edilebilir konumlan-dırma başarımı, düşük enerji tüketimi ve daha az ek yük getirmekte, buna karşılık diğerlerinden da-ha düşük kesinlik sağladığı görülmektedir.

Underwater Sensor Networks (USNs) can improve naval defense, earthquake/tsunami forewarning, wa-ter pollution detection, ocean life monitoring sys-tems, etc. Stationary Underwater Sensor Networks are ideal for securing or monitoring a fixed target region, e.g. monitoring oil drilling platforms for spill detection, harbor entrances for surveillance, ocean bottom for seismic activity observation, etc. On the other hand, mobile untethered Underwater Sensor Networks are flexible and better alternatives for short term exploration of moving targets. For instance, untethered, free-floating underwater sen-sors can track a chemical spill or a pollutant that may be dangerous to human health or sea life. In a sensor network, sensor nodes collect data from their surrounding and tag these data, in order to transmit them to a more powerful node for pro-cessing. Therefore, it is crucial to know the location of the sensor nodes. Location is required for data tagging, as well as, target detection, node tracking, etc. In addition, localization is essential for position-based routing algorithms which are powerful alter-natives to classical routing approaches in Mobile Ad Hoc Networks (MANET). Localization is a well studied topic in terrestrial sensor networks. Nevertheless, in Underwater Sen-sor Networks, localization is still challenging due to several reasons: i) unavailability of the GPS; ii) low bandwidth, long delay and high bit error rate of the acoustic links; iii) necessity of high amount of sen-sor nodes to cover the three dimensional region. The use of GPS is limited to surface nodes because the GPS signal does not propagate through the water. In sensor networking literature, several GPS-less (GPS-free) positioning schemes have been proposed however they usually have high overhead. The un-derwater sensors use acoustic links and the band-width of those links is low even for very short dis-tances. Moreover, acoustic communications has high propagation delay and high bit error rate. In Underwater Sensor Networks, localization protocols are expected to avoid excessive overhead and estab-lish localization with the least possible messages. This is also enforced by the limited battery life of the underwater sensor nodes and the difficulty of re-charging or replacing the batteries in an underwater application. Usually, an underwater application re-quires a large number of sensor nodes because the data rate of the acoustic links increases with de-creasing distance and shorter ranges between nodes, means that more sensor nodes are needed to cover the three dimensional oceanographic zone. In addi-tion, in a mobile Underwater Sensor Network locali-zation should be repeated and stale location infor-mation should be cleared periodically. Considering all these challenges, it is essential to develop novel localization protocols tailored for mobile Underwa-ter Sensor Networks. In this article, we introduce two distributed, scalable localization techniques; Dive and Rise Localization (DNRL) and Proxy Localization (PL). In DNRL, mobile beacons ascend and descend in the water to deliver their GPS driven coordinates. In PL, the al-ready localized nodes act like beacons likewise mul-ti-stage localization which is a preliminary version of PL (Erol et al., 2008). Unlike the previous work, in PL the non-localized nodes use a different metric to choose the best possible proxies among the can-didates which enhances the performance of the pro-tocol. Here, we compare the performance of PL, DNRL and Large-Scale Localization (LSL). LSL is a technique from the literature (Zhou et al., 2007). We evaluate the performance of these schemes in terms of localization success, accuracy, overhead and en-ergy consumption. Since we compare the perfor-mance of the localization techniques for a mobile Underwater Sensor Network, a realistic underwater mobility model is essential. Recently, the works of Caruso et al. (2008) and Erol et al. (2008) have ap-plied the real ocean current behavior to Underwater Sensor Networks. We use the “Meandering Current Mobility with Surface Effect” (MCM-SE) model to compare the performance of the three localization schemes for a mobile underwater sensor network. The main aim of the article is to provide a compari-son between recently proposed localization schemes for Underwater Sensor Networks. Based on the sim-ulation results, we compare and analyze the perfor-mance of three recent methods that are developed for distributed localization in large-scale Underwa-ter Sensor Networks in terms of localization ratio, accuracy, protocol overhead and energy consumption.